6 Efficient Ways to Replace Categorical Values in Pandas DataFrames

Introduction

Hello, I’m a Python enthusiast. A follower asked how to batch‑replace numeric codes (1,2,3,4) in a column with corresponding Chinese words ("开心", "悲伤", "难过", "泪目"). Below are several Pandas solutions.

Solution Overview

The idea is straightforward: use Pandas’ mapping and replacement functions. The original data is created as follows:

df = pd.DataFrame({'col1': [1, 2, 2, 3, 3, 3, 4, 4, 4, 4]})
df

Method 1: map()

Use a dictionary with map to create a new column.

df['col2'] = df['col1'].map({1:"开心", 2:"悲伤", 3:"难过", 4:"泪目"})
df

Method 2: apply() with custom function

Define a function that returns the appropriate string and apply it.

def getValue(s):
    if s==1:
        return '开心'
    elif s==2:
        return '悲伤'
    elif s==3:
        return '难过'
    elif s==4:
        return '泪目'

df['col3'] = df['col1'].apply(getValue)
df

Method 3: replace()

Chain replace calls for each value.

df['col4'] = df['col1'].replace(1, '开心').replace(2, '悲伤').replace(3, '难过').replace(4, '泪目')
df

Method 4: apply() with dictionary lookup

def get_value(s):
    dict = {1:"开心", 2:"悲伤", 3:"难过", 4:"泪目"}
    return dict[s]

df['col5'] = df['col1'].apply(get_value)
df

Method 5: map() with custom function

def get_value(s):
    dict = {1:"开心", 2:"悲伤", 3:"难过", 4:"泪目"}
    return dict[s]

df['col5'] = df['col1'].map(get_value)
df

Method 6: replace() with list mapping

df['col7'] = df['col1'].replace([1, 2, 3, 4], ['开心', '悲伤', '难过', '泪目'])
df

Conclusion

The six approaches—using map, apply, replace, and dictionary look‑ups—demonstrate how to efficiently replace numeric codes with meaningful strings in a Pandas DataFrame, providing flexible options for different coding styles.